An extended analytical framework for heterogeneous implementations of light cryptographic algorithms

Issam W. Damaj*, Hadi Al-Mubasher, Mahmoud Saadeh

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

6 Citations (Scopus)

Abstract

The increased need for data, combined with the emergence of powerful Internet of Things (IoT) devices, has resulted in major security concerns. The decision-making related to choosing an adequate cryptographic algorithm to use is, indeed, an example concern that affects the performance of an implementation. Lightweight or tiny ciphers are considered to be the go-to algorithms when talking about embedded systems and IoT devices. Such ciphers, when properly integrated, are expected to have a minimal effect on the overall device utilization and thus provide effective performance. In this paper, we propose a unified analytical framework for lightweight ciphers as implemented within heterogeneous computing environments. This framework considers a carefully identified set of metrics that can adequately enable the capturing, ranking, and classifying the attained performance. To that end, a designer can make effective evaluations and exact adjustments to an implementation. This framework uses three decision-making approaches, namely the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), Preference Ranking Organization Method for Enrichment Evaluation (PROMETHEE) II, and Fuzzy TOPSIS. Such approaches take into account both hardware and software metrics when deciding on a suitable cryptographic algorithm to adopt. Validation entails a thorough examination and evaluation of several performance classification schemes. The results confirm that the framework is both valid and effective.

Original languageEnglish
Pages (from-to)154-172
Number of pages19
JournalFuture Generation Computer Systems
Volume141
DOIs
Publication statusPublished - 29 Nov 2022

Keywords

  • Algorithms
  • Classification
  • Cryptography
  • Decision making
  • Heterogeneous computing
  • Performance evaluation

Cite this